CN106855853A - Entity relation extraction system based on deep neural network - Google Patents
Entity relation extraction system based on deep neural network Download PDFInfo
- Publication number
- CN106855853A CN106855853A CN201611233056.7A CN201611233056A CN106855853A CN 106855853 A CN106855853 A CN 106855853A CN 201611233056 A CN201611233056 A CN 201611233056A CN 106855853 A CN106855853 A CN 106855853A
- Authority
- CN
- China
- Prior art keywords
- vector
- word
- speech
- entity
- neural networks
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Pending
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/20—Natural language analysis
- G06F40/279—Recognition of textual entities
- G06F40/289—Phrasal analysis, e.g. finite state techniques or chunking
- G06F40/295—Named entity recognition
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/30—Information retrieval; Database structures therefor; File system structures therefor of unstructured textual data
- G06F16/35—Clustering; Classification
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F40/00—Handling natural language data
- G06F40/30—Semantic analysis
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/08—Learning methods
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Health & Medical Sciences (AREA)
- Artificial Intelligence (AREA)
- Audiology, Speech & Language Pathology (AREA)
- Computational Linguistics (AREA)
- General Health & Medical Sciences (AREA)
- Databases & Information Systems (AREA)
- Data Mining & Analysis (AREA)
- Machine Translation (AREA)
Abstract
The present invention relates to natural language processing field, the entity relation extraction system more particularly to based on deep neural network；By in system described in pending text input, the system realizes automatic decision and the output of entity relationship；Part of speech is included characteristic information and is input in convolutional neural networks by the system, completed by convolutional neural networks to including word, part of speech and relative to relation to be extracted provider location information Automatic Feature Extraction, carry out the automatic classification of entity relationship；Without carrying out feature extraction manually, the efficiency and accuracy rate of prediction are higher.The system provides the automatic extraction tool of entity relationship.
Description
Technical field
The present invention relates to natural language processing field, the entity relation extraction system more particularly to based on deep neural network
System.
Background technology
With the fast development of internet, internet has become the main channel that people obtain information, on internet
Text data content also show the trend of exponential growth.Abundant information is contained in text data on internet,
Knowledge base is built for us or knowledge mapping is highly useful；But the artificial relevant knowledge extraction workload that carries out is extremely huge,
If computer it will be appreciated that and extract useful information, that will have very important significance.But the text on internet
Data are nearly all presence, i.e. Un-structured in the form of natural language, and computer cannot be processed directly.In order to solve
This problem, information extraction technique arises at the historic moment, and information extraction technique extracts structuring from the text data of Un-structured
Data, including entity, inter-entity relation, event etc..
Relation extraction is a key technology in information extraction technique, is generally identified by naming entity recognition techniques
Entity in sentence or language, then by the relation between Relation extraction technology identification entity pair.The conventional method bag of Relation extraction
Include：Rule-based abstracting method, the abstracting method based on unsupervised learning and the abstracting method based on supervised learning.Based on rule
Relation extraction method then is manual compiling rule come the relation between two entities in recognizing sentence or language.Based on without prison
The Relation extraction method that educational inspector practises will include that the sentence or language of entity are clustered, and relation knowledge is carried out based on cluster result
Not.Relation extraction method based on supervised learning, is generally converted into a classification problem by Relation extraction problem, then based on tradition
Machine learning techniques or depth learning technology carry out relation classification.
In current Relation extraction method, rule-based method exists clearly disadvantageous, and the method needs manual compiling big
The rule of amount, workload is very big, not easy care and it is necessary to each field redaction rule, it is impossible to expand to well
Other field.Method based on unsupervised learning, when sentence or a language piece are clustered, often effect is not fine, exists and calls together
Rate and preparation rate problem not high are returned, and needs many manual interventions.In Relation extraction algorithm based on supervised learning,
Carried out when relation is classified, it is necessary to manually refine substantial amounts of feature based on traditional machine learning algorithm, it will usually based on participle, word
Property mark and the result of the natural language processing instrument such as syntactic analysis extract feature, and need to have specific domain knowledge,
Workload is big.
When carrying out relation classification based on deep learning, substantial amounts of feature extraction is not done manually, such as utilize convolutional Neural net
Network carries out relation classification, but the part of speech letter for carrying out not having using word in sentence when relation is classified currently with convolutional neural networks
Breath, part of speech is the important achievement of morphological analysis, very meaningful to understanding sentence implication, for relation classification provides very important
Information, it is possible to increase the accuracy rate of relation classification, the effect that optimization relation is extracted.Lack the corresponding entity based on part of speech at present
The automatic extraction tool of relation.
The content of the invention
It is an object of the invention to overcome the above-mentioned deficiency in the presence of prior art, there is provided based on deep neural network
Entity relation extraction system, the system provides the automatic extraction tool of entity relationship, by system described in pending text input
In, the system realizes automatic decision and the output of entity relationship；Part of speech is included characteristic information and is input to convolution by the system
In neutral net, the letter to the provider location including word, part of speech and relative to relation to be extracted is completed by convolutional neural networks
The Automatic Feature Extraction of breath, carries out the automatic classification of entity relationship；Without carry out feature extraction manually, the efficiency of prediction and accurate
Rate is higher.
In order to realize foregoing invention purpose, the invention provides following technical scheme：Entity based on deep neural network
Relation extraction system, the system includes convolutional neural networks, and the system is defeated by the word information vector matrix of pending text
In entering the convolutional neural networks, feature extraction is carried out by the convolutional neural networks, and then complete pending text in pairs
The judgement of entity relationship；
The word information vector matrix is formed by word information vector sequential；
Word information vector from correspondence term vector, part of speech vector, relative to relation first instance to be extracted position to
Measure and be spliced relative to the position vector of second instance.
Specifically, the system realizes that entity relation extraction includes implemented below step：
(1) system carries out participle to pending text, forms word sequence, and each word in sequence is changed to form right
The term vector answered；Corresponding part-of-speech tagging is carried out to each word in sequence, the part of speech of each word is changed into corresponding part of speech
Vector；
(2) each word generates first position vector relative to the position of first instance in calculating sentence；It is each in calculating sentence
Individual word generates second place vector relative to the position of second instance；
(3) by the term vector of each word, part of speech vector in word sequence, first position vector sum second place vector, splicing
Into corresponding word information vector；And by the corresponding word information vector sequential of each word, form word information matrix；
(4) word information matrix is sampled by convolutional neural networks；And then realize that entity relationship classification judges.
Further：The system is also comprising term vector conversion module and part of speech vector modular converter；The term vector turns
Mold changing block, completes term vector conversion and includes implemented below step：
Build a corpus；
Participle is carried out to the text in corpus, and carries out equivalent mark；
Use Word Embedding algorithms to forming word after participle enter row vector conversion, same word correspondence one to
Amount；
Each part of speech is entered into row vector using Word Embedding algorithms to convert, one vector of same part of speech correspondence.
As a kind of preferred：The term vector conversion module and part of speech vector conversion module realize word from word2vec
With the vectorization of part of speech.
Further, the convolutional neural networks include convolutional layer, pond layer and softmax layers；The convolutional layer will be carried
The characteristic information got is input in the layer of pond after carrying out dimension-reduction treatment, be input in softmax layers carry out entity relationship point
Class is predicted.
Further, the system is the computer or server for being loaded with above-mentioned entity relation extraction function program.
Compared with prior art, beneficial effects of the present invention：The present invention provides the entity relationship based on deep neural network
Extraction system, the automatic extraction tool of the offer entity relationship, by system described in pending text input, the system reality
The automatic decision of existing entity relationship and output；Part of speech is included characteristic information and is input in convolutional neural networks by the system, by rolling up
Neutral net is accumulated to complete to being carried including word, part of speech and relative to the automated characterization of the information of the provider location of relation to be extracted
Take, carry out the automatic classification of entity relationship；When Relation extraction is carried out using convolutional neural networks, except utilizing word information and phase
Beyond for the positional information of entity, the part-of-speech information of word is also fully utilized by, helps to be better understood from sentence semantics.By word
Property vector sum term vector is combined into a bigger vector of information content；Part of speech vector information when term vector information is not enough
Convolutional neural networks can be enable to learn automatically to more features for contributing to relation to classify as supplement, accuracy rate is higher.When
Word in sentence represented when there is ambiguity, after adding part-of-speech information, can disambiguation to a certain extent, make Relation extraction
Robustness is more preferable.
Compared with rule-based Relation extraction method, the entity relation extraction that present system is realized is without manual compiling
Substantial amounts of rule, reduces workload；And can be relatively good expand to different fields.Present system is eliminated manually
The step of rule is write, the extraction of feature is completed to extract by convolutional neural networks, compared with based on conventional machines learning method,
The present invention it goes without doing cumbersome Feature Engineering work, makes to improve the judging efficiency of entity relationship.
Brief description of the drawings：
Fig. 1 entity relation extraction systems based on deep neural network that are this realize block diagram.
Fig. 2 is for originally the entity relation extraction system based on deep neural network realizes step schematic diagram in embodiment 1.
Fig. 3 is the operation principle schematic diagram of this entity relation extraction system based on deep neural network.
Specific embodiment
With reference to test example and specific embodiment, the present invention is described in further detail.But this should not be understood
For the scope of above-mentioned theme of the invention is only limitted to following embodiment, all technologies realized based on present invention belong to this
The scope of invention.
Entity relation extraction system based on deep neural network is provided：Entity relation extraction based on deep neural network
System, the system includes convolutional neural networks, and the word information vector matrix of pending text is input into the volume by the system
In product neutral net, feature extraction is carried out by the convolutional neural networks, and then complete to entity relationship in pending text
Judge；
The word information vector matrix is formed by word information vector sequential；
Word information vector from correspondence term vector, part of speech vector, relative to relation first instance to be extracted position to
Measure and be spliced relative to the position vector of second instance.
Specifically, the system realizes that entity relation extraction includes implemented below step：
(1) system carries out participle to pending text, forms word sequence, and each word in sequence is changed to form right
The term vector answered；Corresponding part-of-speech tagging is carried out to each word in sequence, the part of speech of each word is changed into corresponding part of speech
Vector；
(2) each word generates first position vector relative to the position of first instance in calculating sentence；It is each in calculating sentence
Individual word generates second place vector relative to the position of second instance；
(3) by the term vector of each word, part of speech vector in word sequence, first position vector sum second place vector, splicing
Into corresponding word information vector；And by the corresponding word information vector sequential of each word, form word information matrix；
(4) word information matrix is sampled by convolutional neural networks；And then realize that entity relationship classification judges.
Further：The system is also comprising term vector conversion module and part of speech vector conversion module；The term vector turns
Change module, complete term vector conversion and include implemented below step：
Build a corpus；
Participle is carried out to the text in corpus, and carries out equivalent mark；
Use Word Embedding algorithms to forming word after participle enter row vector conversion, same word correspondence one to
Amount；
Each part of speech is entered into row vector using Word Embedding algorithms to convert, one vector of same part of speech correspondence.
As a kind of preferred：The term vector conversion module and part of speech vector conversion module realize word from word2vec
With the vectorization of part of speech.
Further, the convolutional neural networks include convolutional layer, pond layer and softmax layers；The convolutional layer will be carried
The characteristic information got is input in the layer of pond after carrying out dimension-reduction treatment, is carried out to entity relationship in being input to softmax layers
Classification prediction.
Further, the system is the computer or server for being loaded with above-mentioned entity relation extraction function program.
Embodiment 1：
Term vector conversion module and part of speech vector modular converter are set up or stored in computer or server, and are carried out
Training：As shown in Figure 2：Selection one larger corpus, using participle instrument to corpus in all sentences carry out participle,
Obtain word segmentation result.To the word segmentation result of corpus, the N-dimensional term vector of each word is generated using Word Embedding technologies
(size of N latitudes according in corpus comprising word number, i.e. the scale of language material sets；In the larger feelings of corpus
Under condition, in order to avoid encoding sparse problem, dimensionality reduction can be carried out, for example, each word is represented using vector, used in vector
The numeral of continuous change), and then obtain term vector matrix Matrix1 of the language material place comprising word, wherein each row of matrix
A term vector for word in vector correspondence corpus.The same word same vector of correspondence in this step in corpus, such as
Say：" China " one word, may repeatedly occur in corpus, but " China " one word is only corresponding same in vectorial annotation results
Vector.Preferably, word2vec may be selected to realize the vector conversion of each word, word2vec can realize that the vector of word turns
Change, the meaning of a word and semanteme can be better understood from, in the vector for being generated, vector is nearer, and the corresponding meaning of a word is also nearer.
On the basis of participle, the word segmentation result based on corpus, using part-of-speech tagging instrument to each sentence in corpus
Word in son carries out part-of-speech tagging.After the part-of-speech tagging result for obtaining, using the part of speech in sentence an as sequence；Using Word
Embedding technologies generate the M dimension part of speech vectors of each part of speech, and then obtain the part of speech moment of a vector Matrix2 of corpus, wherein
A part of speech vector for part of speech in every a line correspondence corpus of matrix.
Said process establishes the mapping relations of word and term vector, establishes the mapping relations of part of speech and part of speech vector；For
The use of convolutional neural networks is prepared.
The convolutional neural networks model of the system function is built, during above-mentioned model is stored in into computer or server；
Model needs to be trained network after setting up, and training process is as follows：, it is necessary to bag before formal Classification and Identification is carried out
The neutral net for including convolutional layer, maxpooling layer and softmax layers is trained, and training process is similar with identification process；Structure
After building up neutral net, a number of (such as 300) are manually marked the training sample of entity relationship type, carried out
Text participle and part-of-speech tagging, formation sequence search correspondence term vector, in Matrix2 to each word in sequence in Matrix1
It is middle to extract corresponding part of speech vector, generation relative to first instance first position vector, the of generation equivalent to second instance
Two position vectors, a corresponding word information vector is spliced into by the above-mentioned vector of each word, and according to segmentation sequence, by each
The vectorial sequential of word forms matrix, using matrix as convolutional neural networks input, by the forward and reverse propagation of neutral net
With the regulation of automatic weight, when the accuracy rate threshold value for setting, deconditioning.
After model training is finished, by system described in pending text input, the system is automatically obtained mistake identified below
Journey：For the pending text (sentence or language) comprising relationship entity pair to be extracted, participle is carried out using participle instrument, obtained
Sequence after participle, and carry out part-of-speech tagging using part-of-speech tagging instrument.Determine two entity (first instances of sentence to be sorted
Entity1 and second instance Entity2) position, and position of each word relative to first instance Entity1 in the sequence of calculation
Put, to each position position vector Vector3 that one K of generation is tieed up at random, and then obtain position vector matrix Matrix3.
Position of each word relative to second instance Entity2 in the sequence of calculation, to each position position that one K of generation is tieed up at random
Vectorial Vector4, and then obtain position vector matrix Matrix4.
For each word in sentence or language word segmentation result sequence, taken out from term vector matrix Matrix1 successively
Corresponding N-dimensional term vector Vector1；For each word in the part-of-speech tagging result sequence of sentence or language, successively from word
Property vector matrix Matrix2 in take out corresponding M dimension part of speech vectors Vector2.It is word-based relative to Entity1 and Entity2
Position, corresponding K dimension position vector Vector3 is taken out from Matrix3, taken out from Matrix4 corresponding K tie up position to
Amount Vector4, Vector1, Vector2, Vector3 and Vector4 is spliced together and obtains the vector of N+M+2K dimensions
Vector5。
After obtaining the N+M+2K dimensional vectors of each word in sentence or language, sequentially it is grouped together and obtains sentence or language
Vector matrix Matrix5, it is assumed that the length of sequence be C.
Using Matrix5 as the input of convolutional neural networks, convolution operation is done by convolutional layer first, convolution kernel size is
w×(N+M+2K)；The each sliding position of convolution kernel be 1, therefore each convolution kernel can obtain a length for C-w+1 dimension to
Amount.The vector that convolutional layer is gathered is input to pond layer, the latitude of vector is reduced by pond layer, can just use max
Pooling layers is done Max Pooling operations as pond layer to each convolution kernel：Using max pooling functions, that is, take
Go out that of C-w+1 dimensional vectors intermediate value maximum；One convolution kernel correspondence, one value, by each sentence or language after the layer of pond
A piece can obtain number of the dimension of vector Vector6, vectorial Vector6 equal to convolution kernel in network.Finally in pond layer
On the basis of, vectorial Vector6 is input in softmax layers, the classification for carrying out entity relationship with softmax layers is calculated
The class probability of entity relationship in pending text, takes the relationship type of maximum probability as classification results.
As shown in Figure 3：The present embodiment illustrates this hair by taking " XXX groups president king X holds interim board of shareholders " as an example
The extraction process of bright system entity relationship：Enter text into present system, system is automatically performed to " XXX groups president
King X holds interim board of shareholders " participle is carried out, obtain：" XXX groups president king X holds interim board of shareholders " segmentation sequence,
Part-of-speech tagging is carried out to the sequence after participle；Part-of-speech tagging result for " XXX groups/NN president/NN king X/NR holds/VV faces
When/AD boards of shareholders/NN ".Will (by " XXX groups/NN president/NN king X/NR holds/and VV is interim/AD boards of shareholders/NN " it is right
The word information matrix answered) it is input in convolutional layer, the extraction of feature is completed by convolutional neural networks, it is input to max pooling
In layer, dimension-reduction treatment is carried out, finally by the softmax layers of entity of output " XXX groups president king X holds interim board of shareholders "
The other judged result of relation object.
Claims (7)
1. the entity relation extraction system of deep neural network is based on, it is characterised in that the system includes convolutional neural networks,
Be input into the word information vector matrix of pending text in the convolutional neural networks by the system, by the convolutional neural networks
Feature extraction is carried out, and then completes the judgement to entity relationship in pending text；
The word information vector matrix is formed by word information vector sequential；
Word information vector by correspondence term vector, part of speech vector, the position vector relative to relation first instance to be extracted and
Position vector relative to second instance is spliced.
2. the system as claimed in claim 1, it is characterised in that the entity relation extraction of the system includes implemented below step
Suddenly：
(1) system carries out participle to pending text, forms word sequence, and each word in sequence is changed to form corresponding
Term vector；Corresponding part-of-speech tagging is carried out to each word in sequence, the part of speech of each word is changed into corresponding part of speech vector；
(2) each word generates first position vector relative to the position of first instance in calculating sentence；Calculate each word in sentence
Relative to the position of second instance, generation second place vector；
(3) by the term vector of each word, part of speech vector in word sequence, first position vector sum second place vector, it is right to be spliced into
The word information vector answered；And by the corresponding word information vector sequential of each word, form word information matrix；
(4) word information matrix is sampled by convolutional neural networks；And then realize that entity relationship classification judges.
3. system as claimed in claim 2, it is characterised in that：The system is also comprising term vector conversion module and part of speech vector
Modular converter；The term vector modular converter, completes term vector conversion and includes implemented below step：
Build a corpus；
Participle is carried out to the text in corpus, and carries out equivalent mark；
Use Word Embedding algorithms to forming word after participle enter row vector conversion, one vector of same word correspondence；
Each part of speech is entered into row vector using Word Embedding algorithms to convert, one vector of same part of speech correspondence.
4. system as claimed in claim 3, it is characterised in that：The term vector conversion module and part of speech vector modular converter are adopted
The vectorization of word and part of speech is realized with word2vec.
5. system as claimed in claim 4, it is characterised in that：The convolutional neural networks include convolutional layer, pond layer and
Softmax layers；The characteristic information that the convolutional layer will be extracted is input in the layer of pond after carrying out dimension-reduction treatment, is input to
The classification prediction to entity relationship is carried out in softmax layers.
6. system as claimed in claim 5, it is characterised in that：The pond layer is maxpooling layers.
7. the system as described in one of claim 1 to 6, it is characterised in that：The system is taken out to be loaded with above-mentioned entity relationship
Take the computer or server of function program.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611233056.7A CN106855853A (en) | 2016-12-28 | 2016-12-28 | Entity relation extraction system based on deep neural network |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201611233056.7A CN106855853A (en) | 2016-12-28 | 2016-12-28 | Entity relation extraction system based on deep neural network |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN106855853A true CN106855853A (en) | 2017-06-16 |
Family
ID=59126933
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201611233056.7A Pending CN106855853A (en) | 2016-12-28 | 2016-12-28 | Entity relation extraction system based on deep neural network |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN106855853A (en) |
Cited By (37)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN107480122A (en) * | 2017-06-26 | 2017-12-15 | 迈吉客科技（北京）有限公司 | A kind of artificial intelligence exchange method and artificial intelligence interactive device |
| CN107633362A (en) * | 2017-09-18 | 2018-01-26 | 前海梧桐（深圳）数据有限公司 | The method and its system of bind mode between enterprise factor are expressed based on biological characteristic |
| CN107818080A (en) * | 2017-09-22 | 2018-03-20 | 新译信息科技（北京）有限公司 | Term recognition methods and device |
| CN107895051A (en) * | 2017-12-08 | 2018-04-10 | 宏谷信息科技（珠海）有限公司 | A kind of stock news quantization method and system based on artificial intelligence |
| CN107944559A (en) * | 2017-11-24 | 2018-04-20 | 国家计算机网络与信息安全管理中心 | A kind of entity relationship automatic identifying method and system |
| CN107943847A (en) * | 2017-11-02 | 2018-04-20 | 平安科技（深圳）有限公司 | Business connection extracting method, device and storage medium |
| CN108009285A (en) * | 2017-12-22 | 2018-05-08 | 重庆邮电大学 | Forest Ecology man-machine interaction method based on natural language processing |
| CN108133436A (en) * | 2017-11-23 | 2018-06-08 | 科大讯飞股份有限公司 | Automatic method and system of deciding a case |
| CN108280058A (en) * | 2018-01-02 | 2018-07-13 | 中国科学院自动化研究所 | Relation extraction method and apparatus based on intensified learning |
| CN108280061A (en) * | 2018-01-17 | 2018-07-13 | 北京百度网讯科技有限公司 | Text handling method based on ambiguity entity word and device |
| CN108287820A (en) * | 2018-01-12 | 2018-07-17 | 北京神州泰岳软件股份有限公司 | A kind of generation method and device of text representation |
| CN108304911A (en) * | 2018-01-09 | 2018-07-20 | 中国科学院自动化研究所 | Knowledge Extraction Method and system based on Memory Neural Networks and equipment |
| CN108304530A (en) * | 2018-01-26 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Knowledge base entry sorting technique and device, model training method and device |
| CN108416434A (en) * | 2018-02-07 | 2018-08-17 | 复旦大学 | The circuit structure accelerated with full articulamentum for the convolutional layer of neural network |
| CN108595429A (en) * | 2018-04-25 | 2018-09-28 | 杭州闪捷信息科技股份有限公司 | The method for carrying out Text character extraction based on depth convolutional neural networks |
| CN108846017A (en) * | 2018-05-07 | 2018-11-20 | 国家计算机网络与信息安全管理中心 | The end-to-end classification method of extensive newsletter archive based on Bi-GRU and word vector |
| CN109063159A (en) * | 2018-08-13 | 2018-12-21 | 桂林电子科技大学 | A kind of entity relation extraction method neural network based |
| CN109446328A (en) * | 2018-11-02 | 2019-03-08 | 成都四方伟业软件股份有限公司 | A kind of text recognition method, device and its storage medium |
| CN109522557A (en) * | 2018-11-16 | 2019-03-26 | 中山大学 | Training method, device and the readable storage medium storing program for executing of text Relation extraction model |
| CN109661664A (en) * | 2017-06-22 | 2019-04-19 | 腾讯科技（深圳）有限公司 | A kind of method and relevant apparatus of information processing |
| CN109710759A (en) * | 2018-12-17 | 2019-05-03 | 北京百度网讯科技有限公司 | Text dividing method, device, computer equipment and readable storage medium storing program for executing |
| CN109800413A (en) * | 2018-12-11 | 2019-05-24 | 北京百度网讯科技有限公司 | Recognition methods, device, equipment and the readable storage medium storing program for executing of media event |
| CN109885627A (en) * | 2019-02-13 | 2019-06-14 | 北京航空航天大学 | The method and device of relationship between a kind of neural metwork training entity |
| CN109902186A (en) * | 2019-03-12 | 2019-06-18 | 北京百度网讯科技有限公司 | Method and apparatus for generating neural network |
| CN109902309A (en) * | 2018-12-17 | 2019-06-18 | 北京百度网讯科技有限公司 | Interpretation method, device, equipment and storage medium |
| CN110276066A (en) * | 2018-03-16 | 2019-09-24 | 北京国双科技有限公司 | The analysis method and relevant apparatus of entity associated relationship |
| CN110688486A (en) * | 2019-09-26 | 2020-01-14 | 北京明略软件系统有限公司 | Relation classification method and model |
| CN110704547A (en) * | 2019-09-26 | 2020-01-17 | 北京明略软件系统有限公司 | Relation extraction data generation method, model and training method based on neural network |
| CN110928546A (en) * | 2018-09-20 | 2020-03-27 | 西门子股份公司 | Method, apparatus, electronic device, medium, and program for determining existence of dependency violations |
| CN111192692A (en) * | 2020-01-02 | 2020-05-22 | 上海联影智能医疗科技有限公司 | Entity relationship determination method and device, electronic equipment and storage medium |
| CN111274812A (en) * | 2018-12-03 | 2020-06-12 | 阿里巴巴集团控股有限公司 | Character relation recognition method, device and storage medium |
| JP2020107085A (en) * | 2018-12-27 | 2020-07-09 | 楽天株式会社 | Learning device, validity determination device, learning method, validity determination method, learning program, and validity determination program |
| CN111400456A (en) * | 2020-03-20 | 2020-07-10 | 北京百度网讯科技有限公司 | Information recommendation method and device |
| CN112990388A (en) * | 2021-05-17 | 2021-06-18 | 成都数联铭品科技有限公司 | Text clustering method based on concept words |
| CN113076744A (en) * | 2021-04-16 | 2021-07-06 | 重庆中国三峡博物馆 | Cultural relic knowledge relation extraction method based on convolutional neural network |
| CN113792539A (en) * | 2021-09-15 | 2021-12-14 | 平安科技（深圳）有限公司 | Entity relation classification method and device based on artificial intelligence, electronic equipment and medium |
| US11288294B2 (en) | 2018-04-26 | 2022-03-29 | Accenture Global Solutions Limited | Natural language processing and artificial intelligence based search system |
Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106202044A (en) * | 2016-07-07 | 2016-12-07 | 武汉理工大学 | A kind of entity relation extraction method based on deep neural network |
-
2016
- 2016-12-28 CN CN201611233056.7A patent/CN106855853A/en active Pending
Patent Citations (1)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN106202044A (en) * | 2016-07-07 | 2016-12-07 | 武汉理工大学 | A kind of entity relation extraction method based on deep neural network |
Non-Patent Citations (2)
| Title |
|---|
| THIEN HUU NGUYEN 等: "Relation Extraction: Perspective from Convolutional Neural Networks", 《PROCEEDINGS OF NAACL-HLT 2015》 * |
| 洪军建: "面向社会网络应用的人物关系抽取方法研究", 《中国优秀博硕士学位论文全文数据库(硕士) 信息科技辑》 * |
Cited By (57)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| US10789415B2 (en) | 2017-06-22 | 2020-09-29 | Tencent Technology (Shenzhen) Company Limited | Information processing method and related device |
| CN109661664A (en) * | 2017-06-22 | 2019-04-19 | 腾讯科技（深圳）有限公司 | A kind of method and relevant apparatus of information processing |
| WO2019001127A1 (en) * | 2017-06-26 | 2019-01-03 | 迈吉客科技（北京）有限公司 | Virtual character-based artificial intelligence interaction method and artificial intelligence interaction device |
| CN107480122A (en) * | 2017-06-26 | 2017-12-15 | 迈吉客科技（北京）有限公司 | A kind of artificial intelligence exchange method and artificial intelligence interactive device |
| CN107480122B (en) * | 2017-06-26 | 2020-05-08 | 迈吉客科技（北京）有限公司 | Artificial intelligence interaction method and artificial intelligence interaction device |
| CN107633362A (en) * | 2017-09-18 | 2018-01-26 | 前海梧桐（深圳）数据有限公司 | The method and its system of bind mode between enterprise factor are expressed based on biological characteristic |
| CN107633362B (en) * | 2017-09-18 | 2020-11-20 | 前海梧桐（深圳）数据有限公司 | Method and system for expressing connection mode between enterprise elements based on biological characteristics |
| CN107818080A (en) * | 2017-09-22 | 2018-03-20 | 新译信息科技（北京）有限公司 | Term recognition methods and device |
| WO2019085328A1 (en) * | 2017-11-02 | 2019-05-09 | 平安科技（深圳）有限公司 | Enterprise relationship extraction method and device, and storage medium |
| CN107943847A (en) * | 2017-11-02 | 2018-04-20 | 平安科技（深圳）有限公司 | Business connection extracting method, device and storage medium |
| CN107943847B (en) * | 2017-11-02 | 2019-05-17 | 平安科技（深圳）有限公司 | Business connection extracting method, device and storage medium |
| CN108133436A (en) * | 2017-11-23 | 2018-06-08 | 科大讯飞股份有限公司 | Automatic method and system of deciding a case |
| CN107944559A (en) * | 2017-11-24 | 2018-04-20 | 国家计算机网络与信息安全管理中心 | A kind of entity relationship automatic identifying method and system |
| CN107895051A (en) * | 2017-12-08 | 2018-04-10 | 宏谷信息科技（珠海）有限公司 | A kind of stock news quantization method and system based on artificial intelligence |
| CN108009285A (en) * | 2017-12-22 | 2018-05-08 | 重庆邮电大学 | Forest Ecology man-machine interaction method based on natural language processing |
| CN108280058A (en) * | 2018-01-02 | 2018-07-13 | 中国科学院自动化研究所 | Relation extraction method and apparatus based on intensified learning |
| CN108304911A (en) * | 2018-01-09 | 2018-07-20 | 中国科学院自动化研究所 | Knowledge Extraction Method and system based on Memory Neural Networks and equipment |
| CN108287820B (en) * | 2018-01-12 | 2021-06-11 | 鼎富智能科技有限公司 | Text representation generation method and device |
| CN108287820A (en) * | 2018-01-12 | 2018-07-17 | 北京神州泰岳软件股份有限公司 | A kind of generation method and device of text representation |
| CN108280061A (en) * | 2018-01-17 | 2018-07-13 | 北京百度网讯科技有限公司 | Text handling method based on ambiguity entity word and device |
| US11455542B2 (en) | 2018-01-17 | 2022-09-27 | Beijing Baidu Netcom Science And Technology Co., Ltd. | Text processing method and device based on ambiguous entity words |
| CN108280061B (en) * | 2018-01-17 | 2021-10-26 | 北京百度网讯科技有限公司 | Text processing method and device based on ambiguous entity words |
| CN108304530A (en) * | 2018-01-26 | 2018-07-20 | 腾讯科技（深圳）有限公司 | Knowledge base entry sorting technique and device, model training method and device |
| CN108416434B (en) * | 2018-02-07 | 2021-06-04 | 复旦大学 | Circuit structure for accelerating convolutional layer and full-connection layer of neural network |
| CN108416434A (en) * | 2018-02-07 | 2018-08-17 | 复旦大学 | The circuit structure accelerated with full articulamentum for the convolutional layer of neural network |
| CN110276066A (en) * | 2018-03-16 | 2019-09-24 | 北京国双科技有限公司 | The analysis method and relevant apparatus of entity associated relationship |
| CN110276066B (en) * | 2018-03-16 | 2021-07-27 | 北京国双科技有限公司 | Entity association relation analysis method and related device |
| CN108595429A (en) * | 2018-04-25 | 2018-09-28 | 杭州闪捷信息科技股份有限公司 | The method for carrying out Text character extraction based on depth convolutional neural networks |
| US11288294B2 (en) | 2018-04-26 | 2022-03-29 | Accenture Global Solutions Limited | Natural language processing and artificial intelligence based search system |
| CN108846017A (en) * | 2018-05-07 | 2018-11-20 | 国家计算机网络与信息安全管理中心 | The end-to-end classification method of extensive newsletter archive based on Bi-GRU and word vector |
| CN109063159A (en) * | 2018-08-13 | 2018-12-21 | 桂林电子科技大学 | A kind of entity relation extraction method neural network based |
| CN109063159B (en) * | 2018-08-13 | 2021-04-23 | 桂林电子科技大学 | Entity relation extraction method based on neural network |
| CN110928546A (en) * | 2018-09-20 | 2020-03-27 | 西门子股份公司 | Method, apparatus, electronic device, medium, and program for determining existence of dependency violations |
| CN109446328A (en) * | 2018-11-02 | 2019-03-08 | 成都四方伟业软件股份有限公司 | A kind of text recognition method, device and its storage medium |
| CN109522557A (en) * | 2018-11-16 | 2019-03-26 | 中山大学 | Training method, device and the readable storage medium storing program for executing of text Relation extraction model |
| CN109522557B (en) * | 2018-11-16 | 2021-07-16 | 中山大学 | Training method and device of text relation extraction model and readable storage medium |
| CN111274812A (en) * | 2018-12-03 | 2020-06-12 | 阿里巴巴集团控股有限公司 | Character relation recognition method, device and storage medium |
| CN111274812B (en) * | 2018-12-03 | 2023-04-18 | 阿里巴巴集团控股有限公司 | Figure relation recognition method, equipment and storage medium |
| CN109800413A (en) * | 2018-12-11 | 2019-05-24 | 北京百度网讯科技有限公司 | Recognition methods, device, equipment and the readable storage medium storing program for executing of media event |
| CN109902309A (en) * | 2018-12-17 | 2019-06-18 | 北京百度网讯科技有限公司 | Interpretation method, device, equipment and storage medium |
| CN109710759B (en) * | 2018-12-17 | 2021-06-08 | 北京百度网讯科技有限公司 | Text segmentation method and device, computer equipment and readable storage medium |
| CN109902309B (en) * | 2018-12-17 | 2023-06-02 | 北京百度网讯科技有限公司 | Translation method, device, equipment and storage medium |
| CN109710759A (en) * | 2018-12-17 | 2019-05-03 | 北京百度网讯科技有限公司 | Text dividing method, device, computer equipment and readable storage medium storing program for executing |
| JP2020107085A (en) * | 2018-12-27 | 2020-07-09 | 楽天株式会社 | Learning device, validity determination device, learning method, validity determination method, learning program, and validity determination program |
| CN109885627A (en) * | 2019-02-13 | 2019-06-14 | 北京航空航天大学 | The method and device of relationship between a kind of neural metwork training entity |
| CN109902186A (en) * | 2019-03-12 | 2019-06-18 | 北京百度网讯科技有限公司 | Method and apparatus for generating neural network |
| CN110688486A (en) * | 2019-09-26 | 2020-01-14 | 北京明略软件系统有限公司 | Relation classification method and model |
| CN110704547A (en) * | 2019-09-26 | 2020-01-17 | 北京明略软件系统有限公司 | Relation extraction data generation method, model and training method based on neural network |
| CN111192692A (en) * | 2020-01-02 | 2020-05-22 | 上海联影智能医疗科技有限公司 | Entity relationship determination method and device, electronic equipment and storage medium |
| CN111192692B (en) * | 2020-01-02 | 2023-12-08 | 上海联影智能医疗科技有限公司 | Entity relationship determination method and device, electronic equipment and storage medium |
| CN111400456A (en) * | 2020-03-20 | 2020-07-10 | 北京百度网讯科技有限公司 | Information recommendation method and device |
| CN111400456B (en) * | 2020-03-20 | 2023-09-26 | 北京百度网讯科技有限公司 | Information recommendation method and device |
| CN113076744A (en) * | 2021-04-16 | 2021-07-06 | 重庆中国三峡博物馆 | Cultural relic knowledge relation extraction method based on convolutional neural network |
| CN112990388B (en) * | 2021-05-17 | 2021-08-24 | 成都数联铭品科技有限公司 | Text clustering method based on concept words |
| CN112990388A (en) * | 2021-05-17 | 2021-06-18 | 成都数联铭品科技有限公司 | Text clustering method based on concept words |
| CN113792539A (en) * | 2021-09-15 | 2021-12-14 | 平安科技（深圳）有限公司 | Entity relation classification method and device based on artificial intelligence, electronic equipment and medium |
| CN113792539B (en) * | 2021-09-15 | 2024-02-20 | 平安科技（深圳）有限公司 | Entity relationship classification method and device based on artificial intelligence, electronic equipment and medium |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN106855853A (en) | Entity relation extraction system based on deep neural network | |
| CN106649275A (en) | Relation extraction method based on part-of-speech information and convolutional neural network | |
| CN109948165B (en) | Fine granularity emotion polarity prediction method based on mixed attention network | |
| CN110795543B (en) | Unstructured data extraction method, device and storage medium based on deep learning | |
| CN110427463B (en) | Search statement response method and device, server and storage medium | |
| CN106295796B (en) | entity link method based on deep learning | |
| CN106484664B (en) | Similarity calculating method between a kind of short text | |
| CN104834747B (en) | Short text classification method based on convolutional neural networks | |
| CN107463658B (en) | Text classification method and device | |
| CN110083833B (en) | Method for analyzing emotion by jointly embedding Chinese word vector and aspect word vector | |
| CN109558487A (en) | Document Classification Method based on the more attention networks of hierarchy | |
| CN104598611B (en) | The method and system being ranked up to search entry | |
| CN110134946B (en) | Machine reading understanding method for complex data | |
| CN110162636A (en) | Text mood reason recognition methods based on D-LSTM | |
| CN108399158A (en) | Attribute sensibility classification method based on dependency tree and attention mechanism | |
| CN110096711A (en) | The natural language semantic matching method of the concern of the sequence overall situation and local dynamic station concern | |
| CN115393692A (en) | Generation formula pre-training language model-based association text-to-image generation method | |
| CN110321563A (en) | Text emotion analysis method based on mixing monitor model | |
| CN113505200B (en) | Sentence-level Chinese event detection method combined with document key information | |
| CN110377916A (en) | Word prediction technique, device, computer equipment and storage medium | |
| CN111291187B (en) | Emotion analysis method and device, electronic equipment and storage medium | |
| CN107688870A (en) | A kind of the classification factor visual analysis method and device of the deep neural network based on text flow input | |
| CN115455171B (en) | Text video mutual inspection rope and model training method, device, equipment and medium | |
| CN106649250A (en) | Method and device for identifying emotional new words | |
| CN112561718A (en) | Case microblog evaluation object emotion tendency analysis method based on BilSTM weight sharing |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WD01 | Invention patent application deemed withdrawn after publication |
Application publication date: 20170616 |
|
| WD01 | Invention patent application deemed withdrawn after publication |